[llvm] [AMDGPU][Attributor] Add an option to turn on internalization (PR #108420)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,202 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --include-generated-funcs --version 5
+; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -passes=amdgpu-attributor %s | FileCheck --check-prefixes=EXT %s
+; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -passes="amdgpu-attributor<closed-world>" %s | FileCheck --check-prefixes=CW %s
+; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -passes="amdgpu-attributor<internalization>" %s | FileCheck --check-prefixes=INT %s
+
+ at G = global i32 0, align 4
----------------
arsenm wrote:

don't use globals in addrspace(0)

https://github.com/llvm/llvm-project/pull/108420